how does a randomized comparitive experiment control the effects of confounding variables?

Answer 1

I'd say that randomization makes the confounding factors have a net effect of zero. For instance there will be as much negative influence as positive in a random sample so the net effect is zero.